DESCRIPTION: (Applicant's Abstract) Although there is evidence that standard outpatient treatment can be effective for some patients with cocaine dependence, rates of early treatment dropout and relapse remain relatively high. Recent studies have indicated that the addition of services for concurrent psychiatric, interpersonal, and employment problems can improve substance use outcomes for patients, particularly if they have greater problem severity in these areas. This approach is predicated on the idea that professionally-delivered services in these areas help "remove the barriers" to recovery in patients with co-occurring disorders. An independent line of research has generated considerable evidence that providing reinforcement in the form of vouchers, contingent on providing drug-free urine samples that can be redeemed for goods and services can also lead to improved substance use outcomes. These findings suggest that there may be two conceptually and theoretically different, behaviorally-based avenues for improving treatment outcomes for substance abusers. We propose to evaluate these two approaches for cocaine dependent patients who will also receive "treatment as usual" in an outpatient community treatment program. Because of the different purported mechanisms of change in the services matched to problems and voucher conditions, the study will feature a 2 X 2 design to evaluate the potential additive effects of the two interventions. The treatment conditions in the study are as follows: a) Treatment as usual (TAU): b) Voucher reinforcement (TAU+V): c) Services matched to problems (TAU+S) d) Services matched to problems plus vouchers (TAU+SV) The treatment phase of the study will last six months, and patients will be followed up for another year. Both effectiveness and benefit-cost analyses will be conducted. With regard to effectiveness, a main effect for vouchers is predicted during the within-treatment phase. During the post-treatment follow-up, main effects for both the services matched to problems and voucher conditions are hypothesized, along with an interaction effect. Specifically, the TAU+V and TAU+S conditions are hypothesized to produce better cocaine use outcomes than TAU, and the combination of both interventions (TAU+SV) is expected to generate better outcomes than either TAU+V or TAU+ S. With regard to benefit-cost, both the TAU+S and TAU+V interventions are hypothesized to be cost-beneficial, as compared to treatment as usual, and TAU+SV is predicted to be cost-beneficial compare to TAU+S and TAU+V. Analyses will also compare the benefit-cost of professional services vs. voucher reinforcement. Additional analyses will identify predictors of response to the TAU+V and TAU+S conditions, and test predictions regarding possible mechanisms of action.